{"id":18439475,"url":"https://github.com/idiap/nnsslm","last_synced_at":"2025-04-07T21:32:32.862Z","repository":{"id":60451868,"uuid":"185949341","full_name":"idiap/nnsslm","owner":"idiap","description":"Neural Network based Sound Source Localization Models","archived":false,"fork":false,"pushed_at":"2023-08-29T04:55:28.000Z","size":3841,"stargazers_count":38,"open_issues_count":1,"forks_count":9,"subscribers_count":5,"default_branch":"master","last_synced_at":"2025-03-23T01:01:53.351Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"bsd-3-clause","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/idiap.png","metadata":{"files":{"readme":"README.rst","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2019-05-10T08:24:10.000Z","updated_at":"2025-03-14T09:46:23.000Z","dependencies_parsed_at":"2024-11-06T06:37:41.588Z","dependency_job_id":null,"html_url":"https://github.com/idiap/nnsslm","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/idiap%2Fnnsslm","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/idiap%2Fnnsslm/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/idiap%2Fnnsslm/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/idiap%2Fnnsslm/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/idiap","download_url":"https://codeload.github.com/idiap/nnsslm/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":247732694,"owners_count":20986904,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-11-06T06:24:55.208Z","updated_at":"2025-04-07T21:32:27.851Z","avatar_url":"https://github.com/idiap.png","language":"Python","funding_links":[],"categories":["Tools"],"sub_categories":["SSL+"],"readme":"Neural Network based Sound Source Localization Models\n=====================================================\n\nThis repository includes the programs to run and test the neural network models that we proposed in our publications on sound source localization, including\n\n* Deep learning based direction-of-arrival (DOA) estimation\n* Joint DOA estimation and speech/non-speech classification\n\nThe networks has been trained to work with the microphone array of the Softbank Pepper robot (early version with directional microphones).\n\nPlease cite the relevant publications when using the code.\n\n\nDependency\n----------\n\n* `Python \u003chttps://www.python.org/\u003e`_ (version \u003e= 3.8)\n* `NumPy \u003chttp://www.numpy.org/\u003e`_ (version \u003e= 1.14)\n* `PyTorch \u003chttps://pytorch.org/\u003e`_ (version \u003e= 0.2)\n* `apkit \u003chttps://github.com/hwp/apkit\u003e`_ (version \u003e= 0.2)\n\n\nData\n----\n\nWe use the `SSLR dataset \u003chttps://www.idiap.ch/dataset/sslr\u003e`_ for the experiments.\n\n\nUsage\n-----\n\nTraining\n********\n\nThis command reproduces the \"ResNet-STFT\" model described in `\"Deep Learning Approaches for Auditory Perception in Robotics\", Weipeng He, PhD Thesis, EPFL \u003chttps://infoscience.epfl.ch/record/283940\u003e`_:\n\n.. code::\n\n  ssl_nn_v2/train_resnet_2stage.py --model=models/test --window-size=8192 --hop-size=4096 --n-doa=360 --output-act=5 --n-out-hidden=1 --sigma=8 --s1-epoch=4 --epoch=10 --lr=0.001 --ld=2 --batch-size=128 \u003cSSLR_PATH\u003e/lsp_train_*\n\nInference\n*********\n\n.. code::\n\n  ssl_nn_v2/test_nn_raw.py --feature=stft --method=METHOD_NAME --window-size=8192 --hop-size=4096 --batch-size=100 \u003cSSLR_PATH\u003e/human models/thesis_resnet_act5_p1lsp_s1ep4_ep10_valid_b100\n\nEvaluation\n**********\n\n.. code::\n\n  eval/gen_2tasks_report.py --method=METHOD_NAME --window-size=8192 --hop-size=4096 --output=REPORT_DIR --ssl-only \u003cSSLR_PATH\u003e/human\n  gnuplot --persist REPORT_DIR/ssl_pr_plot\n\nPublications\n------------\n\nThe models and code in this repository are based on the work published in:\n\n  Deep Neural Networks for Multiple Speaker Detection and Localization\n  Weipeng He, Petr Motlicek, Jean-Marc Odobez \n  In *IEEE International Conference on Robotics and Automation (ICRA)*, 2018\n\n  Joint Localization and Classification of Multiple Sound Sources Using a Multi-task Neural Network\n  Weipeng He, Petr Motlicek, Jean-Marc Odobez \n  In *INTERSPEECH*, 2018\n\n  Deep Learning Approaches for Auditory Perception in Robotics\n  Weipeng He\n  PhD Thesis, EPFL\n\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fidiap%2Fnnsslm","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fidiap%2Fnnsslm","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fidiap%2Fnnsslm/lists"}